Gram-negative bacterial sepsis and shock remains a significant cause of morbidity and mortality in surgical patients. Increasing evidence has indicated that gram-negative bacteremia is associated with concurrent endotoxemia, and that even low level endotoxemia can provoke a markedly exaggerated host response in which macrophage activation clearly plays an early pivotal role. Gram-negative lipopolysaccharide (LPS, endotoxin) is perhaps the most potent stimulus of macrophage monokine production [tumor necrosis factor (TNF), interleukin-1 (IL-1), and IL-6]. Although the lipid A portion of LPS has been associated in experimental models with the majority of toxic effects, the host humoral response to gram-negative bacteremia or endotoxemia is largely serotype specific and is not directed against the lipid A moiety. The deep core/lipid A (DCLA) region of LPS, however, represents a highly biochemically and immunologically conserved region of the LPS molecule, and is thus an ideal candidate against which cross-reactive antibody with anti-toxin activity could be developed. Several problems, however, exist at present: 1) the exact binding site within the DCLA region that will serve to maximize cross-reactivity and protection has not been established, 2) the mechanism by which anti-DCLA monoclonal antibodies (mAbs) bind to intact bacteria or LPS in vitro and in vivo has not been defined, 3) in vitro assays that will serve as predictors of in vivo protective capacity have not been developed, and 4) the mechanism by which anti-LPS mAbs provide protection in vivo has not been defined. The objective of this research program, therefore, is to examine and define the immunologic and microbiologic basis of the ability of anti- LPS mAbs to afford protection during experimental gram-negative bacterial sepsis. The DCLA antigenic region is expressed extensively on the cell surface of rough mutants of Escherichia coli and Salmonella minnesota that thus represent suitable immunogens for cross-reactive antibody production. Previous work on this project has confirmed the capacity of anti-core LPS mAbs to cross-react in vitro and provide protection in vivo. The major thrust of this work will therefore be to develop and study anti-DCLA mAbs. Different classes of mAbs with similar binding specificity will be developed via in vitro selection of class switch clones. These mAbs will be extensively characterized in vitro for: 1) antibody titer, 2) LPS binding, 3) bacterial binding specificity, 4) opsonophagocytic character, and 5) inhibition of macrophage monokine secretion. In vivo testing in acute and chronic peritonitis models will serve to: 1) to establish protective capacity, 2) to correlate in vitro characterization with protective capacity and via this correlation to develop screening assays for rapid selection of protective mAbs, and 3) to establish the immunologic and microbiologic basis through which anti-DCLA mAbs serve to promote survival during experimental gram-negative bacterial sepsis.